Professional Documents
Culture Documents
Alexander Christopher Notes On The Synthesis of Form PDF
Alexander Christopher Notes On The Synthesis of Form PDF
Alexander Christopher Notes On The Synthesis of Form PDF
�
t
I
1
t
•
NOTES ON THE SYNTHESIS
\
Today, almost ten years after I wrote this book, one idea stands out
clearly for me as the most important in the book: the idea of the diagrams.
These diagrams, which, in my more recent work, I. have been calling
patterns, are the key to the process of creating form. In this book I pre
sented the diagrams as the end results of a long process; I put the accent
on the process, and gave the diagrams themselves only a few pages of
discussion. But once the book was finished, and I began to explore the
process which I had described, I found that the diagrams themselves
had immense power, and that, in fact, most of the power of what I had
written lay in the power of these diagrams.
The idea of a diagram, or pattern, is very simple. It is an abstract pat
tern of physical relationships which resolves a small system of interacting
and conflicting forces, and is independent of all other forces, and of all
other possible diagrams. The idea that it is possible to create such ab
stract relationships one at a time, and to create designs which are whole
by fusing these relationships-this amazingly simple idea is, for me, the
most important discovery of the book.
I have discovered, since, that these abstract diagrams not only allow
you to create a single whole from them, by fusion, but also have other
even more important powers. Because the diagrams are .independent of
one another, you can study them and improve them one at a time, so
that their evolution can be gradual and cumulative. More important still,
because they are abstract and independent, you can use them to create
not just one design, but an infinite variety of designs, all of them free
combinations of the same set of patterns.
As you can see, it is the independence of the diagrams which gives them
these powers. At the time I wrote this book, I was very much concerned
with the formal definition of "independence," and the idea of using a
mathematical method to discover systems of forces and diagrams which
are independent. But once the book was written, I discovered that it is quite
unnecessary to use such a complicated and formal way of getting at the inde
pendent diagrams.
If you understand the need to create independent diagrams, which re-
solve, or solve, systems of interacting human forces, you will find that
you can create, and develop, these diagrams piecemeal, one at a time, in
the most natural way, out of your experience of buildings and design,
simply by thinking about the forces which occur there and the conflicts
between these forces.
I have written about this realization and its consequences, in other,
more recent works. But I feel it is important to say it also here, to make
you alive to it before you read the book, since so many readers have
focused on the method which leads to the creation of the diagrams, not on
the diagrams themselves, and have even made a cult of following this
method.
Indeed, since the book was published, a whole academic field has
grown up around the idea of "design methods"-and I have been hailed
as one of the leading exponents of these so-called design methods. I am
very sorry that this has happened, and want to state, publicly, that I
reject the whole idea of design methods as a subject of study, since I
think it is absurd to separate the study of designing from the practice of
design. In fact, people who study design methods without also practicing
design are almost always frustrated designers who have no sap in them,
who have lost, or never had, the urge to shape things. Such a person will
never be able to say anything sensible about "how" to shape things
either.
Poincare once said: "Sociologists discuss sociological methods; physi
cists discuss physics." I love this statement. Study of method by itself is
always barren, and people who have treated this book as if it were a book
about "design method" have almost always missed the point of the
diagrams, and their great importance, because they have been obsessed
with the details of the method I propose for getting at the diagrams.
No one will become a better designer by blindly following this method,
or indeed by following any method blindly. On the other hand, if you
try to understand the idea that you can create abstract patterns by
studying the implication of limited systems of forces, and can create new
forms by free combination of these patterns-and realize that this will
only work if the patterns which you define deal with systems of forces
whose internal interaction is very dense, and whose interaction with the
other forces in the world is very weak-then, in the process of trying to
create such diagrams or patterns for yourself, you will reach the central
idea which this book is all about.
C.A.
Berkeley, California
February 197 1
CONTENTS
I. Introduction:
The Need for Rationality
Part One
2. Goodness of Fit I5
Part Two
6. The Program 73
8. De finit ions 95
9· Solution I I6
Epilogue I32
Notes I93
"First, the taking in of scattered particulars under one Idea,
so that everyone understands what is being talked about ... Sec
ond, the separation of the Idea into parts, by dividing it at the
joints, as nature directs, not breaking any limb in half as a bad
carver might."
simplicity
j ointing
economy
This is a typical design problem ; it has requirements which
have to be met ; and there are interactions between the re
quirements, which makes the requirements hard to meet. This
problem is simple to solve. It falls easily within the compass
of a single man's intuition. But what about a more compli
cated problem?
Consider the task of designing a complete environment for
a million people. The ecological balance of human and animal
and plant life must be correctly adjusted both internally and
to the given exterior physical conditions. People must be able
to lead the individual lives they wish for. The social conditions
induced must not lead to gross ill-health or to gross personal
misery, and must not cause criminal delinquency. The cyclical
intake of food and goods must not interfere with the regular
movements of the inhabitants. The economic forces which
2
develop must not lead to real-estate speculation which de
stroys the functional relation between residential areas and
areas supporting heavy goods. The transportation system must
not be organized so that it creates a demand that aggravates
its own congestion. People must somehow be able to live in
close cooperation and yet pursue the most enormous variety
of interests. The physical layout must be compatible with
foreseeable future regional developments. The conflict be
tween population growth and diminishing water resources,
energy resources, parklands, must somehow be taken care of.
The environment must be organized so that its own regener
ation and reconstruction does not constantly disrupt its
performance.
As in the simpler example, each of these issues interacts
with several of the others. But in this case each issue is itself
a vast problem ; and the pattern of interactions is vastly com
plicated. The difference between these two cases is really like
the difference between the problem of adding two and two,
and the problem of calculating the seventh root of a fifty digit
number. In the first case we can quite easily do it in our
heads. In the second case, the complexity of the problem will
defeat us unless we find a simple way of writing it down,
which lets us break it into smaller problems.
Today more and more design problems are reaching in
soluble levels of complexity. This is true not only of moon
bases, factories, and radio receivers, whose complexity is
internal, but even of villages and teakettles. In spite of their
superficial simplicity, even these problems have a background
of needs and activities which is becoming too complex to
grasp intuitively.
To match the growing complexity of problems, there is a
3
growing body of information and specialist experience. This
information is hard to handle ; it is widespread, diffuse, un
organized.1 Moreover, not only is the quantity of information
itself by now beyond the reach of single designers, but the
various specialists who retail it are narrow and unfamiliar
with the form-makers' peculiar problems, so that it is never
clear quite how the designer should best consult them.2 As
a result, although ideally a form should reflect all the known
facts relevant to its design, in-fact the average designer scans
whatever information he happens on, consults a consultant
now and then when faced by extra-special difficulties, and
introduces this randomly selected information into forms
otherwise dreamt up in the artist's studio of his mind. The
technical difficulties of grasping all the information needed for
the construction of such a form are out of hand - and well
beyond the fingers of a single individual.3
At the same time that the problems increase in quantity,
complexity, and difficulty, they also change faster than before.
New materials are developed all the time, social patterns alter
quickly, the culture itself is changing faster than it has ever
changed before. In the past - even after the intellectual
upheaval of the Renaissance - the individual designer would
stand to some extent upon the shoulders of his predecessors.
And although he was expected to make more and more of his
own decisions as traditions gradually dissolved, there was
always still some body of tradition which made his decisions
easier. Now the last shreds of tradition are being torn from
him. Since cultural pressures change so fast, any slow de
velopment of form becomes impossible. Bewildered, the form
maker stands alone. He has to make clearly conceived forms
without the possibility of trial and error over time. He has
4
to be encouraged now to think his task through from the
beginning, and to " create " the form he is concerned with, for
what once took many generations of gradual development is
now attempted by a single individual.4 But the burden of a
thousand years falls heavily on one man's shoulders, and this
burden has not yet materially been lightened. The intuitive
resolution of contemporary design problems simply lies be
yond a single individual's integrative grasp .
Of course there are no definite limits to this grasp ( especially
in view of the rare cases where an exceptional talent breaks
all bounds) . But if we look at the lack of organization and lack
of clarity of the forms around us, it is plain that their design
has often taxed their designer's cognitive capacity well beyond
the limit. The idea that the capacity of man's invention is
limited is not so surprising, after all. In other areas it has been
shown , and we admit readily enough, that there are bounds
to man's cognitive and creative capacity. There are limits to
the difficulty of a laboratory problem which he can solve ; 5 to
the number of issues he can consider simultaneously ; 6 to the
complexity of a decision he can handle wisely.7 There are no
absolute limits in any of these cases ( or usually even any scale
on which such limits could be specified ) ; yet in practice it is
clear that there are limits of some sort. Similarly, the very
frequent failure of individual designers to produce well or
ganized forms suggests strongly that there are limits to the
individual designer's capacity.
II
PART ONE
2 I GOODNESS O F FIT
38
passes through is a record or history of the adaptation be
tween form and context. The history of the system displays
the form-making process at work. To compare unselfconscious
and selfconscious form-making processes, we have only to
examine the kinds of history which the system of variables
can have in these two processes. As we shall see, the kinds of
history which the system can have in the unselfconscious and
selfconscious processes are very different.
We shall perhaps understand the idea of a system's history
best if we make a simple picture of it.28
Imagine a system of a hundred lights. Each light can be
in one of two possible states. In one state the light is on. The
lights are so constructed that any light which is on always
has a 50-50 chance of going off in the next second. In the other
state the light is off. Connections between lights are con
structed so that any light which is off has a 50-50 chance of
going on again in the next second, provided at least one of the
lights it is connected to is on. If the lights it is directly con
nected to are off, for the time being it has no chance of going
on again, and stays off. If the lights are ever all off simultane
ously, then they will all stay off for good, since when no light
is on, none of the lights has any chance of being reactivated.
This is a state of equilibrium. Sooner or later the system of
lights will reach it.
This system of lights will help us understand the history of
a form-making process. Each light is a binary variable, and
so may be thought of as a misfit variable. The off state cor
responds to fit ; the on state corresponds to misfit. The fact
that a light which is on has a 50-50 chance of going off
every second, corresponds to the fact that whenever a mis
fit occurs efforts are made to correct it. The fact that lights
39
which are off can be turned on again by connected lights,
corresponds to the fact that even well-fitting aspects of
a form can be unhinged by changes initiated to correct
some other misfit because of connections between variables.
The state of equilibrium, when all the lights are off, corre
sponds to perfect fit or adaptation. It is the equilibrium in
which all the misfit variables take the value 0. Sooner or later
the system of lights will always reach this equilibrium. The
only question that remains is, how long will it take for this
to happen? It is not hard to see that apart from chance this
depends only on the pattern of interconnections between the
lights.
Let us consider two extreme circumstances.29
1. On the one hand, suppose there are no interconnections
between lights at all. In this case there is nothing to prevent
each light's staying off for good, as soon as it goes off. The
average time it takes for all the lights to go off is therefore
only a little greater than the average time it takes for a single
light to go off, namely 21 seconds or 2 seconds.
2. On the other hand, imagine such rich interconnections
between lights that any one light still on quickly rouses all
others from the off state and puts them on again. The only
way in which this system can reach adaptation is by the pure
chance that all 100 happen to go off at the same moment.
The average time which must elapse before this happens will
be of the order of 2100 seconds, or 1022 years.
The second case is useless. The age of the universe itself is
only about 1010 years. For all intents and purposes the system
will never adapt. But the first case is no use either. In any
real system there are interconnections between variables
which make it impossible for each variable to adapt in com-
40
plete isolation. Let us therefore construct a third possibility.
3. In this case suppose there are again interconnection s
among the 100 lights, but that we discern in the pattern of
interconnection s some 10 principal subsystems, each contain
ing 10 lights.30 The lights within each subsystem are so strongly
connected to one another that again all 10 must go off simul
taneously before they will stay off ; yet at the same time the
subsystems themselves are independent of one another as
wholes, so that the lights in one subsystem can be switched
off without being reactivated by others flashing in other sub
systems. The average time it will take for all 100 lights to go
off is about the same as the time it takes for one subsystem to
go off, namely 210 seconds, or about a quarter of an hour.
Of course, real systems do not behave so simply. But fifteen
minutes is not much greater than the two seconds it takes an
isolated variable to adapt, and the enormous gap between
these magnitudes and 1022 years does teach us a vital lesson.
No complex adaptive system will succeed in adapting in a
reasonable amount of time unless the adaptation can proceed
subsystem by subsystem, each subsystem relatively inde
pendent of the others.31
This is a familiar fact. It finds a close analogy in the chil
dren's sealed glass-fronted puzzles which are such fun and so
infuriating. The problem, in these puzzles, is to achieve cer
tain configurations within the box : rings on sticks, balls in
sockets, pieces of various shapes in odd-shaped frames - but
all to be done by gentle tapping on the outside of the box.
Think of the simplest of these puzzles, where half a dozen
colored beads, say, are each to be put in a hole of corres
ponding color.
One way to go about this problem would be to pick the
4 1
puzzle up, give it a single energetic shake, and lay it down
again, in the hope that the correct configuration would appear
by accident. This ali-or-nothing method might be repeated
many thousand times, but it is clear that its chances of success
are negligible. It is the technique of a child who does not under
stand how best to play. Much the easiest way - and the way
we do in fact adopt under such circumstances - is to juggle
one bead at a time. Once a bead is in, provided we tap gently,
it is in for good, and we are free to manipulate the next one
that presents itself, and we achieve the full configuration step
by step . When we treat each bead as an isolable subsystem,
and take the subsystems independently, we can solve the
puzzle.
If we now consider the process of form-making, in the light
of these examples, we see an easy way to make explicit the
distinction between processes which work and those which
don't.
Let us remind ourselves of the precise sense in which there
is a system active in a form-making process. It is a purely
fictitious system. Its variables are the conditions which must
be met by good fit between form and context. Its interactions
are the causal linkages which connect the variables to one
another. If there is not enough light in a house, for instance,
and more windows are added to correct this failure, the change
may improve the light but allow too little privacy ; another
change for more light makes the windows bigger, perhaps,
but thereby makes the house more likely to collapse. These
are examples of inter-variable linkage. If we represent this
system by drawing a point for each misfit variable, and a link
between two points for each such causal linkage, we get a
structure which looks something like this:
42
Now, let us go back to the question of adaptation. Clearly
these misfit variables, being interconnected, cannot adjust
independently, one by one . On the other hand, since not all
the variables are equally strongly connected (in other words
there are not only dependences among the variables, but also
independences) , there will always be subsystems like those
circled below, which can, in principle, operate fairly inde
pendently. 3 2
Now when there are such irritations, how fast does the fail
ure lead to action, how quickly does it lead to a change of
form? Think first, perhaps, of man's closeness to the ground in
the unselfconscious culture, and of the materials he uses when
48
he makes his house. The Hebridean crofter uses stone and
clay and sods and grass and straw, all from the near surround
ings.l5 The Indian's tent used to be made of hide from the
buffalo he ate.16 The Apulian uses as building stones the very
rocks which he has taken from the ground to make his agri
culture possibleP These men have a highly developed eye for
the trees and stones and animals which contain the means of
their livelihood, their food, their medicine, their furniture,
their tools. To an African tribesman the materials available
are not simply objects, but are full of life.18 He knows them
through and through; and they are always close to hand.
Closely associated with this immediacy is the fact that the
owner is his own builder, that the form-maker not only makes
the form but lives in it.Indeed, not only is the man who lives
in the form the one who made it, but there is a special close
ness of contact between man and form which leads to constant
rearrangement of unsatisfactory detail, constant improvement.
The man, already responsible for the original shaping of the
form, is also alive to its demands while he inhabits it.19 And
anything which needs to be changed is changed at once.
The Abipon, whose dwelling was the simplest tent made of
two poles and a mat, dug a trench to carry off the rain if it
bothered him.20 The Eskimo reacts constantly to every change
in temperature inside the igloo by opening holes or closing
them with lumps of snow.21 The very special directness of
these actions may be made clearer, possibly, as follows.Think
of the moment when the melting snow dripping from the roof
is no longer bearable, and the man goes to do something
about it. He makes a hole which lets some cold air in, per
haps. The man realizes that he has to do something about
it- but he does not do so by remembering the general rule
49
and then applying it ("When the snow starts to melt it is too
hot inside the igloo and therefore time to ..." ) . He simply
does it. And though words may accompany his action, they
play no essential part in it.This is the important point. The
failure or inadequacy of the form leads directly to the action .
This directness is the second crucial feature of the unself
conscious system's form-production. Failure and correction
go side by side. There is no deliberation in between the
recognition of a failure and the reaction to it.22The directness
is enhanced, too, by the fact that building and repair are so
much an everyday affair.The Eskimo, on winter hunts, makes
a new igloo every night.23 The Indian's tepee cover rarely
lasts more than a single season.24 The mud walls of the Tal
lensi hut need frequent daubs.25 Even the elaborate communal
dwellings of the Amazon tribes are abandoned every two or
three years, and new ones built.26 Impermanent materials and
unsettled ways of life demand constant reconstruction and
repair, with the result that the shaping of form is a task
perpetually before the dweller's eyes and hands. If a form is
made the same way several times over, or even simply left
unchanged, we can be fairly sure that its inhabitant finds
little wrong with it.Since its materials are close to hand, and
their use his own responsibility, he will not hesitate to act
if there are any minor changes which seem worth making.
Let us return now to the question of adaptation.The basic
principle of adaptation depends on the simple fact that the
process toward equilibrium is irreversible. Misfit provides an
incentive to change; good fit provides none. In theory the
process is eventually bound to reach the equilibrium of well
fitting forms.
However, for the fit to occur in practice, one vital condi-
so
tion must be satisfied. It must have time to happen . The
process must be able to achieve its equilibrium before the
next culture change upsets it again . It must actually have
time to reach its equilibrium every time it is disturbed- or,
i f w e see the process a s continuous rather than intermittent,
the adjustment of forms must proceed more quickly than the
drift of the culture context. Unless this condition is fulfilled
the system can never produce well-fitting forms, for the
equilibrium of the adaptation will not be sustained.
As we saw in Chapter 3 , the speed of adaptation depends
essentially on whether the adaptation can take place in in
dependent and restricted subsystems, or not. Although we
cannot actually see these subsystems in the unselfconscious
process, we can infer their activity from the very two char
acteristics of the process which we have been discussing:
directness and tradition .
The direct response is the feedback of the process.27 If the
process is to maintain the good fit of dwelling forms while
the culture drifts, it needs a feedback sensitive enough to
take action the moment that one of the potential failures
actually occurs. The vital feature of the feedback is its im
mediacy. For only through prompt action can it prevent the
build-up of multiple failures which would then demand simul
taneous correction - a task which might, as we have seen,
take too long to be feasible in practice.
However, the sensitivity of feedback is not in itself enough
to lead to equilibrium. The feedback must be controlled, or
damped, somehow. 28 Such control is provided by the resistance
to change the unselfconscious culture has built into its tradi
tions. We might say of these traditions, possibly, that they
make the system viscous. This viscosity damps the changes
5 I
made, and prevents their extension to other aspects of the
form . As a result only urgent changes are allowed. Once a
form fits well, changes are not made again until it fails to
fit again. Without this action of tradition , the repercussions
and ripples started by the slightest failure could grow wider
and wider until they were spreading too fast to be corrected .
On the one hand the directness of the response to misfit
ensures that each failure is corrected as soon as it occurs,
and thereby restricts the change to one subsystem at a time.
And on the other hand the force of tradition, by resisting
needless change, holds steady all the variables not in the
relevant subsystem, and prevents those minor disturbances
outside the subsystem from taking hold . Rigid tradition and
immediate action may seem contradictory. But it is the very
contrast between these two which makes the process self
adjusting. It is just the fast reaction to single failures, com
plemented by resistance to all other change, which allows
the process to make series of minor adjustments instead of
spasmodic global ones: it is able to adjust subsystem by
subsystem, so that the process of adjustment is faster than
the rate at which the culture changes; equilibrium is certain
to be re-established whenever slight disturbances occur ; and
the forms are not simply well-fitted to their cultures, but in
active equilibrium with them.29
54
5 I THE SELFCONSCIOUS PROCESS
s6
right, and is not to be explained simply by the passing of
the unselfconscious process.
I do not wish to imply here that there is any unique process
of development that makes selfconscious culcures out of un
selfconscious ones. Let us remember anyway that the dis
tinction between the two is artificial. And, besides, the facts
of history suggest that the development from one to the other
can happen in rather different ways.1 From the point of view
of my present argument it is immaterial how the development
occurs. All that matters, actually, is that sooner or later the
phenomenon of the master craftsman takes control of the
form-making activities.
One example, of an early kind, of developing selfcon
sciousness is found in Samoa. Although ordinary Samoan
houses are built by their inhabitants-to-be, custom demands
that guest houses be built exclusively by carpenters.2 Since
these carpenters need to find clients, they are in business as
artists; and they begin to make personal innovations and
changes for no reason except that prospective clients will
judge their work for its inventiveness. 3
The form-maker's assertion of his individuality is an im
portant feature of selfconsciousness. Think of the willful
forms of our own limelight-bound architects. The individual,
since his livelihood depends on the reputation he achieves,
is anxious to distinguish himself from his fellow architects,
to make innovations, and to be a star.4
The development of architectural individualism is the clear
est manifestation of the moment when architecture first turns
into a selfconscious discipline. And the selfconscious archi
tect's individualism is not entirely willful either. It is a natural
57
consequence of a man's decision to devote his life exclusively
to the one activity called " architecture. " 5 Clearly it is at
this stage too that the activity first becomes ripe for serious
thought and theory. Then, with architecture once established
as a discipline, and the individual architect established, entire
institutions are soon devoted exclusively to the study and
development of design. The academies are formed. As the
academies develop , the unformulated precepts of tradition
give way to clearly formulated concepts whose very formula
tion invites criticism and debate. 6 Question leads to unrest,
architectural freedom to further selfconsciousness, until it
turns out that (for the moment anyway) the form-maker's
freedom has been dearly bought. For the discovery of archi
tecture as an independent discipline costs the form-making
process many fundamental changes. Indeed, in the sense I
shall now try to describe, architecture did actually fail from
the very moment of its incepti on . With the invention of a
teachable discipline called "architecture, " the old process of
making form was adulterated and its chances of success de
stroyed.
The source of this trouble lies with the individual. In the
unselfconscious system the individual is no more than an
agent.7 He does what he knows how to do as best he can.
Very little demand is made of him. He need not himself be
able to invent forms at all. All that is required is that he
should recognize misfits and respond to them by making
minor changes. It is not even necessary that these changes
be for the better. As we have seen, the system, being self
adjusting, finds its own equilibrium - provided only that
misfit incites some reaction in the craftsman. The forms pro
duced in such a system are not the work of individuals, and
ss
their success does not depend on any one man's artistry, but
only on the artist's place within the process.8
The selfconscious process is different. The artist's self
conscious recognition of his individuality has deep effect on
the process of form-making. Each form is now seen as the
work of a single man, and its success is his achievement only.
Selfconsciousness brings with it the desire to break loose, the
taste for individual expression, the escape from tradition and
taboo, the will to self-determination. But the wildness of the
desire is tempered by man's limited invention. To achieve in a
few hours at the drawing board what once took centuries of
adaptation and development, to invent a form suddenly which
clearly fits its context- the extent of the invention neces
sary is beyond the average designer.
A man who sets out to achieve this adaptation in a single
leap is not unlike the child who shakes his glass-topped puzzle
fretfully, expecting at one shake to arrange the bits inside
correctly.9 The designer's attempt is hardly random as the
child's is ; but the difficulties are the same. His chances of
success are small because the number of factors which must fall
simultaneously into place is so enormous.
Now, in a sense, the limited capacity of the individual
designer makes further treatment of the failure of selfcon
sciousness superfluous. If the selfconscious culture relies on
the individual to produce its forms, and the individual isn't
up to it, there seems nothing more to say. But it is not so
simple. The individual is not merely weak. The moment he
becomes aware of his own weakness in the face of the enormous
challenge of a new design problem, he takes steps to overcome
his weakness ; and strangely enough these steps themselves
exert a very positive bad influence on the way he develops
59
forms. In fact, we shall see that the selfconscious system's
lack of success really doesn't lie so much in the individual's
lack of capacity as in the kind of efforts he makes, when he is
selfconscious, to overcome this incapacity.
6o
bring home the amorphous nature of design problems as they
present themselves to the designer.Naturally the design of a
complex object like a motor car is much more difficult and
requires a much longer list.It is hardly necessary to speculate
as to the length and apparent disorder of a list which could
adequately define the problem of designing a complete urban
environment.
How is a designer to deal with this highly amorphous and
diffuse condition of the problem as it confronts him? What
would any of us do?
Since we cannot refer to the list in full each time we think
about the problem, we invent a shorthand notation. We
classify the items, and then think about the names of the
classes: since there are fewer of these, we can think about
them much more easily. To put it in the language of psy
chology, there are limits on the number of distinct concepts
which we can manipulate cognitively at any one time, and
we. are therefore forced, if we wish to get a view of the whole
problem, to re-encode these items.10 Thus, in the case of the
kettle, we might think about the class of requirements gen
erated by the process of the kettle's manufacture, its capacity,
its safety requirements, the economics of heating water, and
its good looks. Each of these concepts is a general name for
a number of the specific requirements. If we were in a very
great hurry (or for some reason wanted to simplify the
problem even further), we might even classify these concepts
in turn, and deal with the problem simply in terms of (1) its
function and (2) its economics. In this case we would have
erected a four-level hierarchy like that in the diagram on the
next page.
By erecting such a hierarchy of concepts for himself, the
61
a
kettle
�A
production safety use capital maintenance
designer is, after all, able to face the problem all at once.
He achieves a powerful economy of thought, and can by this
means thread his way through far more difficult problems
than he could cope with otherwise. If hierarchies seem less
common in practice than I seem to suggest, we have only to
look at the contents of any engineering manual or architects'
catalogue; the hierarchy of chapter headings and subheadings
is organized the way it is, precisely for cognitive convenience.U
To help himself overcome the difficulties of complexity,
the designer tries to organize his problem. He classifies its
various aspects, thereby gives it shape, and makes it easier to
handle. What bothers him is not only the difficulty of the
problem either. The constant burden of decision which he
comes across, once freed from tradition, is a tiring one. So
he avoids it where he can by using rules ( or general principles) ,
which he formulates in terms of his invented concepts. These
principles are at the root of all so-called "theories" of archi
tectural design.12 They are prescriptions which relieve the
burden of selfconsciousness and of too much responsibility.
62
It is rash, perhaps, to call the invention of either concepts
or prescriptions a conscious attempt to simplify problems. In
practice they unfold as the natural outcome of critical dis
cussion about design. In other words, the generation of verbal
concepts and rules need not only be seen abstractly as the
supposed result of the individual's predicament, but may be
observed wherever the kind of formal education we have
called selfconscious occurs.
A novice in the unselfconscious situation learns by being
put right whenever he goes wrong. " No, not that way, this
way . " No attempt is made to formulate abstractry just what
the right way involves. The right way is the residue when all
the wrong ways are eradicated. But in an intellectual atmos
phere free from the inhibition of tradition , the picture changes.
The moment the student is free to question what he is told,
and value is put on explanation, it becomes important to
decide why "this " is the right way rather than " that , " and
to look for general reasons. Attempts are made to aggregate
the specific failures and successes which occur, into principles.
And each such general principle now takes the place of many
separate and specific admonitions. It tells us to avoid this
kind of form, perhaps, or praises that kind. With failure and
success defined, the training of the architect develops rapidly.
The huge list of specific misfits which can occur, too complex
for the student to absorb abstractly and for that reason usu
ally to be grasped only through direct experience, as it is
in the unselfconscious culture, can now be learned - because
it has been given form. The misfit variables are patterned
into categories like " economics " or " acoustics. " And con
densed, like this, they can be taught, discussed, and criticized.
It is this point, where these concept-determined principles
63
begin to figure in the training and practice of the architect,
that the ill-effect of selfconsciousness on form begins to show
itself.
73
that occur in economics, checkers, logic, or administration,
which can be clarified and solved mechanically.1 They can be
solved mechanically, because they are well enough under
stood for us to turn them into selection problems. 2
To solve a problem by selection, two things are necessary.
74
mechanical way of telling, purely from the drawings which
describe it, whether or not it meets its requirements. Either
we must put the real thing in the actual world, and see
whether it works or not, or we must use our imagination and
experience of the world to predict from the drawings whether
it will work or not. But there is no general symbolic connec
tion between the requirements and the form's description
which provide criteria ; and so there is no way of testing the
form symbolically. 3 Third, even if these first two objections
could be overcome somehow, there is a much more conclusive
difficulty. This is the same difficulty, precisely, that we come
across in trying to construct scientific hypotheses from a
given body of data. The data alone are not enough to define
a hypothesis ; the construction of hypotheses demands the
further introduction of principles like simplicity (Occam's
razor) , non-arbitrariness, and clear organization.4 The con
struct ion of form, too, requires these principles. There is at
present no prospect of introducing these principles mechani
cally, either into science or into design. Again, they require
invention.
It is therefore not possible to replace the actions of a trained
designer by mechanically computed decisions. Yet at the
same time the individual designer's inventive capacity is too
limited for him to solve design problems successfully entirely
by himself. If theory cannot be expected to invent form, how
is it likely to be useful to a designer?
75
context form
actual world
context form
ml El actual world
+ t
IEl -- • lEI mental picture
context form
ml El actual world
• t
IE] lEI mental picture
+ t
ffil • • lEI formal picture of
mental picture
The first scheme represents the unselfconscious situation
described in Chapter 4. Here the process which shapes the
form is a complex two-directional interaction between the
context Cl and the form F l , in the world itself. The human
being is only present as an agent in this process. He reacts to
misfits by changing them ; but is unlikely to impose any
" designed " conception on the form.
The second scheme represents the selfconscious situation
described in Chapter 5. Here the design process is remote
from the ensemble itself ; form is shaped not by interaction
between the actual context's demands and the actual inade
quacies of the form, but by a conceptual interaction between
the conceptual picture of the context which the designer has
learned and invented, on the one hand, and ideas and dia
grams and drawings which stand for forms, on the other. This
interaction contains both the probing in which the designer
searches the problem for its major " issues, " and the develop
ment of forms which satisfy them ; but its exact nature is
unclear. 5 In present design practice, this critical step, during
which the problem is prepared and translated into design,
always depends on some kind of intuition. Though design is
by nature imaginative and intuitive, and we could easily
trust it if the designer's intuition were reliable, as it is it
inspires very little confidence.
In the unselfconscious process there is no possibility of mis
construing the situation : nobody makes a picture of the con
text, so the picture cannot be wrong. But the selfconscious
designer works entirely from the picture in his mind, and this
picture is almost always wrong.
The way to improve this is to make a further abstract
picture of our first picture of the problem, which eradicates
77
its bias and retains only its abstract structural features ; this
second picture may then be examined according to precisely
defined operations, in a way not subject to the bias of language
and experience. 6 The third scheme in the diagram represents a
third process, based on the use of such a picture. The vague
and unsatisfactory picture of the context's demands, C2,
which first develops in the designer's mind, is followed by
this mathematical picture, C3. Similarly, but in reverse, the
design F2 is preceded by an orderly complex of diagrams
F3 . The derivation of these diagrams F3 from C 3 , though
still intuitive, may be clearly understood. The form is actu
ally shaped now by a process at the third level, remote from
<;2 or F2. It is out in the open, and therefore under control.
1 . . . m) . 9
The great power and beauty of the set, as an analytical tool
for design problems, is that its elements can be as various as
they need be, and do not have to be restricted only to require
ments which can be expressed in quantifiable form. Thus in
the design of a house, the set M may contain the need for
individual solitude, the need for rapid construction, the need
for family comfort, the need for easy maintenance, as well as
such easily quantifiable requirements as the need for low
capital cost and efficiency of operation. Indeed, M may con
tain any requirement at all.
79
These requirements are the individual conditions which
must be met at the form-context boundary, in order to prevent
misfit. The field structure of this form-context boundary, in
so far as the designer is aware of it, is also not hard to describe.
He knows that some of the misfits interfere with one another,
as he tries to solve them, or conflict ; that others have common
physical implications, or concur ; and that still others do not
interact at all. It is the presence and absence of these inter
actions which give the set M the system character already
referred to in Chapters 3, 4, and 5 .1 0 We represent the inter
actions by associating with M a second set L, of non-directed,
signed, one-dimensional elements called links, where each link
joins two elements of M, and contains no other elements of M.
As we shall see in Chapter 8, the links bear a negative sign
if they indicate conflict, and a positive sign if they indicate
concurrence, and may also be weighted to indicate strength
of interaction.
The two sets M and L together define a structure known
as a linear graph or topological l -complex, which we shall refer
to as G(M,L) , or simply G for short.U A typical graph is shown
below. Such a graph serves as a picture of a designer's view of
8o
some specific problem. It is a fairly good picture, in the sense
that its constituents, the sets M and L, are available to him
introspectively without too much trouble ; also because it
keeps our attention, neatly and abstractly, on the fact that the
set of misfits has a structure, or, as we called it in Chapter 2,
a field. 1 2
94
8 I DEFINITIONS
From two variables we get four sets, in which the forms take
values as shown below.
10 4
If we superimpose all m variables, we get a division of the
domain D into 2m mutually exclusive classes, each labeled by
a different pattern of values for x1 Xm . We shall call the• • •
metrical. Thus
X a
0 2 0
2 0 -1
X a 0 -1 0
I II
From this matrix we define the set L as a set of links asso
ciated with the variables of M, as follows.16 For every pair
of variables Xi and xh there are I Vii I distinct elements of L
which join Xi to xi. These elements bear the same sign as the
index Vih negative for conflict, and positive for concurrenceY
The sets M and L together, completely define the graph
G(M,L).18
The definitions we have given so far still leave certain
practical questions about the sets M andL unanswered. Does
it matter, for instance, if two variables are very close in
meaning, though slightly different? How specific or how gen
eral must they be? What do we do about three-variable
interaction? The answers to these questions depend on three
important formal properties of the system G(M,L), which we
shall now explore.
First of all, if the graph G(M,L) is to give us an accurate
picture of the variables' behavior, it is necessary that the set
L describe all the interaction between variables which there is.
Since the elements ofL are links which represent two-variable
correlation, this means that the variables must be chosen to
be free from three-variable and higher-order correlations. The
mathematics of Appendix 2 is also based on the assumption
that the higher-order correlations vanish.19 If this is not so,
any analysis based on MandL alone is sure to give misleading
results.
Second, even the two-variable correlation ovii must be
small, for each pair of variables. Specifically, as far as the
mathematics of Appendix 2 is concerned, we must have
lo <:: 1, where l is the total number of links in L.20
Third, the analysis in Appendix 2 is also based on the
assumption of a certain simple symmetry among the variables
I I 2
of M. It demands that p(x; 0) should be the same for all i.21
=
I I 3
In practice, of course, the preciseness of this mathematical
expression is meaningless, since we judge the correlations
"by eye," and do not obtain them numerically. What it does
mean, in practice, though, is that we must be satisfied that all
the variables are as independent as we can get them to be.
An example should make this clear: Suppose the following two
variables appear on our list, for the kettle problem.
1. "The kettle must heat water fast enough."
2. "The kettle must keep water hot once it is heated."
These two are clearly not at all independent. However, there
are two fairly independent issues lurking behind them, if we
can only find them. One way to bring this out would be by
the following rearrangement, which covers more or less the
same ground as the first pair, but consists of two more in
dependent variables.
3. "The kettle must permit one-way heat transmission
only."
4. "The kettle must have low thermal capacity."
A considerable amount of energy must be spent in the pre
liminary stages shuffling and reshuffling the variables in this
fashion, until they are as independent as they can be made. 23
The first formal property, that the three-variable or higher
order correlations among the elements of M should be neg
ligible, is the hardest of all to achieve. It means that the
two-variable correlation for any pair of variables must be
independent of the states of all other variables. Since the
state of one variable is most likely to affect the correlation
between other variables, if that one variable is wide in scope
the best we can do in satisfying this is to make all the indi
vidual variables as specific and minute as possible.
I I 4
This policy of making all the variables highly specific is
important for another reason. However much we may try to
steer clear of existing categories, in practice we shall always
have to generate the specific variables of M through interme
diate stages. The brain is not made to think of such detailed
lists amorphously. Whether we like it or not, if we think of one
variable which has to do with acoustics, we shall inevitably
then think of others which seem, to us, to fall under the
same heading or to be in the same conceptual area. It is
therefore a matter of practical psychology that we cannot
avoid using superordinate concepts like "economics" and
"acoustics" altogether, as intermediate steps in the task of
listing misfit variables. At best we may treat these conceptual
intermediates as key words, as loosely conceived labels for the
principal issues in the problem, which we shall then break
down further into finer pieces to get our set of variables M.
The closer our variables are to these abstract and general key
words, the more susceptible the problem remains to the kind
of distortions discussed in Chapter 5. The more specific and
detailed we make the variables, the less constrained G(M,L)
will be by previous conceptions, and the more open to detailed
and unbiased examination of its causal structure.
I I 5
9 I SOLUTION
I I8
First we must be able to find constructive diagrams for S1
and S2 individually. This means that the misfits which S1
contains must cohere somehow, and suggest a physical aspect
or component of the form under consideration; and the same
for s2.
Secondly, if the decomposition is to serve any useful pur
pose, it must not be necessary to construct the diagram for S3
from scratch. Instead, it must be possible to derive a con
structive diagram for Sa in some simple way from the diagrams
already constructed for sl and s2 in isolation.
To put it simply, the first of these conditions depends on
the internal structure of the sets sl and s2, while the second
deals with the relations between these two sets.
Let us take the two conditions in order.
!20
especially easy to diagram, and do not really " belong " to
the problem.
If we are to make anything sensible of the subsets in this
program, we must now ask just which sets of points to con
sider as being the most " diagrammable . " This depends on the
pattern of interactions between the misfits. Where, after all,
does the interaction among the requirements spring from? It
springs from the intractable nature of the available materials
and the conditions under which the form has to be made. Two
misfits are seen to interact only because, in some sense at
least, they deal with the same kind of physical consideration.
If they dealt with utterly different aspects, there could be no
basis either for conflict or for concurrence.
In building, the need for acoustic insulation conflicts with
the need to build with easily transportable prefabricated ma
terials. These two needs conflict because the first calls for
massive inert walls, while the second calls for light walls. The
physical feature of the world their interaction depends on is
mass. Again, in a highway, the need for safety on curves con
flicts with the need to keep land costs down, because the
wider the curves have to be for safety's sake, the larger the
area eaten up by the transition curves at interchanges. In
this case the interaction between the two requirements de
pends on the radius of the curve.
It is such a physical center of implication, if I may call it
that, which the designer finds it easy to grasp. Because it
refers to a distinguishable physical property or entity, it can
be expressed diagrammatically, and provides a possible non
verbal point of entry into the problem. If we can find sets of
variables in which there are specially dense interactions, we
I2 I
may assume, in these cases, that the density of the interaction
resides in a particularly strong identifiable physical aspect of
the problem. These sets will be the easiest of all to grasp
constructively. Thus:
a
M
I 2 7
to see for every S some verbal paradigm like " This one
deals with the acoustic aspects of the form." If he tries
to do that, he denies the whole purpose of the analysis,
by allowing verbal preconceptions to interfere with the
pattern which the program shows him. The effect of the
design program is that each set of requirements draws
his attention to just one major physical and functional
issue, rather than to some verbal or preconceived issue.
It thereby forces him to consolidate the physical ideas
present in his mind as seedlings, and to make physical
order out of them.
I 3 I
EPILOGUE
My main task has been to show that there is a deep and important
underlying structural correspondence between the pattern of a
problem and the process of designing a physical form which
answers that problem. I believe that the great architect has in the
past always been aware of the patterned similarity of problem and
process, and that it is only the sense of this similarity of structure
that ever led him to the design of great forms.
The same pattern is implicit in the action of the unself
conscious form-producing system, and responsible for its success.
But before we can ourselves turn a problem into form, because we
are selfconscious, we need to make explicit maps of the problem's
structure, and therefore need first to invent a conceptual framework
for such maps. This is all that I have tried to do.
Since my effort may well meet with resistance, I like to see the
few steps taken here reflected in a parable of an imaginary past
society.
Suppose there was once a people who had no formalized arith
metic. When they wanted what we think of as arithmetical results,
they got them by guessing. So if they wished to know the area of a
corn patch they paced its two sides (six paces by ten paces, say) ,
and then mulled the two numbers over. Eventually one of them
came up with an answer - he would say some number, that
is, which estimated the bags of corn needed to sow that patch.
He might say 60, 61, 58, whatever occurred to him. (If we
were in such a situation we should form what we call the prod-
1 3 2
uct of the two numbers, 60, and determine the amount of corn
needed in terms of this area.)
It is easy to see that the people of this imaginary society might
not have found formal arithmetic acceptable. Their own method
was usually not too jar off the mark (sowing corn is such a loose
test, anyway, that what we call inaccuracy would not have been
noticeable) - and besides, there was something rather noble about
the seers (magicians?) who performed the tasks of "calculation. "
Some men were better a t i t than others, certainly; some had the
power to produce appropriate answers, some produced answers
rather wider of the mark. But that didn't seem to matter. Instead
the power was regarded as a great human gift, the people who
possessed it were honored for their capability. A nd both these seers
themselves and their admirers opposed the introduction of a
formalized arithmetic most rigidly, did not see the possible develop
ments, were interested only in preserving their own limited
capacities for calculation.
Such resistance was not altogether foolish either. There were
wise men, too, among those who opposed arithmetic. They foresaw,
correctly, the materialism which it would induce. Its very first
achievement, once introduced, would be to make calculation more
precise and easier, and thereby to save corn. A nd soon number and
economy and size would dominate the human being. The immedi
ate good done by the formulation of arithmetic would be small, and
not worth taking risks for on its own account.
What neither the wise men nor the seers foresaw, however, was
the miraculous developments that this formulation later led to.
By first understanding the shape of the technique which produced
the form of the result, man found further insight. He found that
it is not only the result which is important, but the process too.
Not only the form of the results, but the form of the path which
led to them. It was only by questioning the foundations of
geometry and the processes of geometrical proof that Riemann
1 33
invented the geometry which later became the basis for Einstein's
theory of relativity. Other great theorems are possible today be
cause multiplication and addition were once defined. It was only
because man gave thought to the seemingly obvious processes which
underlay arithmetic that he was able to refine mathematics, and
able to proceed to forms of still higher order, mathematical shapes
of greater elegance and fuller understanding.
The shapes of mathematics are abstract, of course, and the
shapes of architecture concrete and human. But that difference
is inessential. The crucial quality of shape, no matter of what
kind, lies in its organization, and when we think of it this way we
call it form. Man's feeling for mathematical form was able to
develop only from his feeling for the processes of proof. I believe
that our feeling for architectural form can never reach a compa
rable order of development, until we too have first learned a com
parable feeling for the process of design.
1 34
APP E ND I C E S
APPENDIX I I A WORKED EX A M P L E
Here is a worked example, taken from a recent paper, " The De
termination of Components for an Indian Village. " * The problem
treated is this. An agricultural village of six hundred people is
to be reorganized to make it fit present and future conditions
developing in rural India .
The set M, which follows, contains all the misfit variables that
are pertinent to the organization of the village. All these misfit
variables are stated here in their positive form ; that is, as needs
or requirements which must be satisfied positively in a properly
functioning village. They are, however, all derived from state
ments about potential misfits : each one represents some aspect
of the village which could go wrong, and is therefore a misfit
variable in the terms of Chapter 2 .
M includes variables which represent three different kinds of
need :
( 1 ) all those which are explicitly felt by villagers themselves
as needs,
(2) all those which are called for by national and regional
economy and social purpose, and
(3) all · those already satisfied implicitly in the present vil
lage (which are required, though not felt as needs by anybody) .
(The headings on the left are for convenience in the listing stage
only, and play no part in the subsequent analysis.)
* In Christopher Jones , ed. , Conference on Design Method (Oxford :
Pergamon , 1 963) . My lists and diagrams are rep roduced here by kind
p ermission of Pergamon Press .
Religion and Caste
1. Harijans regarded as ritually impure, untouchable, etc.
2. Proper disposal of dead.
3. Rules about house door not facing south .
4. Certain water and certain trees are thought of as sacred.
5. Provision for festivals and religious meetings.
6. Wish for temples.
7 . Cattle treated as sacred, and vegetarian attitude.
8 . Members of castes maintain their caste profession as far as
possible.
9. Members of one caste like to be together and separate from
others, and will not eat or drink together.
1 0 . Need for elaborate weddings.
Social Forces
11. Marriage is to person from another village.
12. Extended family is in one house.
1 3 . Family solidarity and neighborliness even after separation.
14. Economic integration of village on payment-in-kind basis.
1 5 . Modern move toward payment in cash.
1 6 . Women gossip extensively while bathing, fetching water, on
way to field latrines, etc.
17. Village has fixed men's social groups.
1 8 . Need to divide land among sons of successive generations.
1 9 . People want to own land personally.
20. People of different factions prefer to have no contact.
2 1 . Eradication of untouchability.
22. Abolition of Zamindari and uneven land distribution.
23. Men's groups chatting, smoking, even late at night.
24. Place for village events - dancing, plays, singing, etc. , wres
tling.
25. Assistance for physically handicapped, aged, widows.
26. Sentimental system : wish not to destroy old way of life ;
love of present habits governing bathing, food, etc.
13 7
27. Family is authoritarian.
28. Proper boundaries of ownership and maintenance responsibil
ity.
29. Provision for daily bath, segregated by sex, caste, and age.
A griculture
30. Efficient and rapid distribution of seeds, fertilizer, etc . , from
block HQ.
3 1 . Efficient distribution of fertilizer, manure, seed, from village
storage to fields.
32. Reclamation and use of uncultivated land.
33. Fertile land to be used to best advantage.
34. Full collection of natural manure (animal and human) .
35. Protection of crops from insects, weeds, disease.
36. Protection of crops from thieves, cattle, goats, monkeys, etc.
37 . Provision of storage for distributing and marketing crops.
38. Provision of threshing floor and its protection from marauders.
39. Best cotton and cash crop.
40. Best food grain crop.
4 1 . Good vegetable crop.
42. Efficient plowing, weeding, harvesting, leveling.
43. Consolidation of land.
44. Crops must be brought home from fields.
45. Development of horticulture.
46. Respect for traditional agricultural practices.
47 . Need for new implements when old ones are damaged, etc.
48. Scarcity of land.
49. Cooperative farming.
A nimal Husbandry
Water
67. Drinking water to be good, sweet.
68. Easy access to drinking water.
69. Fullest possible irrigation benefit derived from available water.
70. Full collection of underground water for irrigation.
7 1 . Full collection of monsoon water for use.
7 2 . Prevent famine if monsoon fails.
7 3 . Conservation of water resources for future.
7 4 . Maintenance of irrigation facilities.
7 5 . Drainage of land to prevent waterlogging, etc.
76. Flood control to protect houses, roads, etc.
Material Welfare
77. Village and individual houses must be protected from fire.
78. Shade for sitting and walking.
79. Provision of cool breeze.
8 0 . Security for cattle.
139
81. Security for women and children.
82. Provision for children to play ( under supervision) .
83. In summer people sleep in open.
84. Accommodat ion for panchayat records, meetings, etc.
85. Everyone's accommodati on for sitting and sleeping should be
protected from rain.
86. No overcrowding.
87 . Safe storage of goods.
88. Place to wash and dry clothes.
89. Provision of goods, for sale.
90. Better provision for preparing meals.
91. Provision and storage of fuel.
92. House has to be cleaned, washed, drained.
93. Lighting.
Transportation
94. Provision for animal traffic.
95. Ac�ess to bus as near as possible.
96. Access to railway station.
97. Minimize transportation costs for bulk produce ( grain, pota
toes, etc. ) .
98. Daily produce requires cheap and constant ( monsoon ) access
to market.
99. Industry requires strong transportation support.
1 00 . Provision for bicycle age in every village by 1 965.
101. Pedestrian traffic within village.
102. Accommodation for processions.
103. Bullock cart access to house for bulk of grain, fodder.
Forests and Soils
1 04 . Plant ecology to be kept healthy.
105. Insufficient forest land .
1 06 . Young trees need protection from goats, etc.
107. Soil conservation.
108. Road and dwelling erosion.
1 09 . Reclamation of eroded land , gullies, etc.
1 10. Prevent land erosion.
1 40
Education
1 1 1 . Provision for primary education.
1 1 2 . Access to a secondary school.
1 13. Good attendance in school.
1 14. Development of women's independent activities.
1 1 5. Opportunity for youth activities.
1 1 6. Improvement of adult literacy.
1 17 . Spread of information about birth control, disease, etc.
1 18. Demonstration projects which spread by example.
1 19. Efficient use of school ; no distraction of students.
Health
1 20 . Curative measures for disease available to villagers.
121. Facilities for birth, pre- and post-natal care, birth control.
122. Disposal of human excreta.
123. Prevent breeding germs and disease starters.
124. Prevent spread of human disease by carriers, infection, con
tagion.
125. Prevent malnutrition.
Implementation
126. Close contact with village-level worker.
1 27 . Contact with block development officer and extension officers.
128. Price assurance for crops.
1 29 . Factions refuse to cooperate or agree.
1 30 . Need for increased incentives and aspirations.
131. Panchayat must have more power and respect.
1 32 . Need to develop projects which benefit from government
subsidies.
the signs of the links are not indicated : as we shall see in Appen
dix 2 , the decomposition turns out to be independent of the link
signs. The table below simply shows those linked pairs of vari
ables for which v ;; 1 or - 1 . =
L:::..
ENTIRE VILLAGE
A 8 c D
����
At A2 A3 81 82 83 84 C1 C2 D1 D2 D3
A2
A3
A3: 37 Provision of storage for distributing and marketing crops.
38 Provision of threshing floor and its protection from ma-
rauders.
50 Protected storage of fodder.
55 Cattle access to water.
77 Village and individual houses must be protected from fire.
91 Provision and storage of fuel.
103 Bullock cart access to house for bulk of grain, fodder.
Access for cattle to water (55) should be to good water, hence
to drinking water distribution system, feeding off compound
wall D2. 77 and 91 are best achieved by a controlled fuel supply,
like gas, supplied by a gober gas plant using manure from A2,
the gas distributed to individual kitchens by the same artery that
distributes water, i.e., the compound wall.
At the point on the compound wall indicated by these previous
items, there must be an opening to allow passage of bullock carts
(103), and at this point there should also be a store for supplies
and fodder- or at least an easy unloading and access point to
the roofs of the cattle bays (37, 38, 50).
81
B 1: 39 Best cotton and cash crop.
40 Best food grain crop.
41 Good vegetable crop.
44 Crops must be brought home from fields.
51 Improve quality of fodder available.
118 Demonstration projects which spread by example.
1 27 Contact with block development officer.
131 Panchayat must have more power and respect.
138 Achieve economic independence so as not to strain na
tional transportation and resources.
39, 40, 41, 51, and economic independence (138) are all items
which can only be improved by the widespread use of improved
agricultural methods; these are not directly dependent on the
physical plan, but on a change of attitude in the villagers. This
change of attitude cannot be brought about by sporadic visits
from the agricultural extension officer and village-level worker,
but only by the continuing presence of demonstration methods,
on site (118); there should be a demonstration farm, government
or panchayat-owned ( 131), perhaps run by the village-level worker
in association with the panchayat (hence accommodation for such
officers, 127) . 118 and 44 suggest that the farm be placed in such a
way that every farmer passes it daily, on his way to and from the
fields.
r6 o
82
84
B4 : a� Reclamation and use of uncultivated land.
45 Development of horticulture.
48 Scarcity of land.
70 Full collection of underground water for irrigation.
71 Full collection of monsoon water for use.
73 Conservation of water resources for future.
75 Drainage of land to prevent waterlogging, etc.
104 Plant ecology to be kept healthy.
105 Insufficient forest land.
108 Road and dwelling erosion.
109 Reclamation of eroded land, gullies, etc.
32 and 48 call for use of wasteland, which often contains river
bed area. 48 calls for irrigation of this area. 71, 73, 75, suggest
the use of monsoon water instead of and as well as well water for
irrigation, since well irrigation is temporary in the long run, be
cause it causes a drop in the water table. Apart from actually
using monsoon water for irrigation, the water table in the wells
can be preserved if the wells are backed up by a tank. Hence a
curved bund collects water above wells placed under the bund (70) .
Rainfall in the catchment area (again a water resource issue, 73)
will be improved by tree planting (104, 105) , which suggests put
ting fruit trees (45) inside the curve of the bund. (Incidentally,
placing the trees within the bund offers us a way of protecting
young trees from cattle, by keeping the cattle on the other side
of the bund, which then forms a natural barrier.) Further, if water
is to flow toward the tank, horizontal contour bunds cannot
be used to check erosion as they are in B3, so erosion of gullies,
streams , etc., can only be controlled by tree planting (109) . Road
_
erosion is controlled if the road is on top of the bund itself (108) .
8
C1: 8 Members of castes maintain their caste profession as far
as possible.
10 Need for elaborate weddings.
11 Marriage is to person from another village.
14 Economic integration of village on payment-in-kind basis.
15 Modern move toward payment in cash.
58 Development of other animal industry.
63 Development of village industry.
64 Simplify the mobility of labor, to and from villages,
and to and from fields and industries and houses.
65 Diversification of village's economic base- not all occu-
pations agricultural .
66 Efficient provision and use of power.
93 Lighting.
95 Access to bus as near as possible.
96 Access to railway station.
99 Industry requires strong transportation support.
100 Provision for bicycle age in every village by 1965.
112 Access to a secondary school .
121 Facilities for birth, pre- and post-natal care (birth con
trol ) .
130 Need for increased incentives and aspirations.
I 65
132 Need to develop projects which benefit from govern
ment subsidies.
133 Social integration with neighboring villages.
134 Wish to keep up with achievements of neighboring vil
lages.
139 Proper connection with bridges, roads, hospitals, schools,
proposed at the district level.
141 Prevent migration of young people and harijans to
cities.
This is composed of two major functional sets: 11, 64, 95, 100,
112, 121, 133, 134, 139, which concerns the integration of the village
with neighboring villages and with the region, and 8, 10, 14, 15,
58, 63, 65, 66, 93, 96, 99, 130, 132, 141, which concerns the future
economic base of the village, and all the aspects of "modern" life
and society.
These two are almost inseparable. They call for a center, away
from the heart of the village, on the road, able, because of being
on the road, to sustain connections between the village and other
villages (11) and capable of acting as a meeting place for villagers
of different villages (112, 121) . This function is promoted by the
need to provide a bus stop ( 95) , village industries with optimum
access to the road (63-66, 99) , the social gathering place connected
with the bus and with jobs made available by the industries
( 61, 133, 134) ; the development of a modern and almost urban
atmosphere to combat migration of the best people to cities (141 ) ,
and to develop incentives (14, 15, 130, 132) . A center of industry
promotes 8, 63, 64. The road satisfies 64, 95, 96, 99, 100, 139.
The center will be the natural physical location for sources of power
and electricity transformer (66, 93) ; also the most efficient place
for the poultry and dairy farming which require- road access ( 58) ;
the bus stop is the natural arrival place for incoming wedding pro
cessions (10) .
!6 6
C2: 5 Provision for festivals and religious meetings.
6 Wish for temples.
20 People of different factions prefer to have no contact.
21 Eradication of untouchability.
24 Place for village events-dancing, plays, singing, etc.,
wrestling.
84 Accommodation for panchayat records, meetings, etc.
89 Provision of goods, for sale.
102 Accommodation for processions.
1 1 1 Provision for primary education.
1 15 Opportunity for youth activities.
1 16 Improvement of adult literacy.
1 17 Spread of information about birth control, disease, etc.
120 Curative measures for disease available to villagers.
129 Factions refuse to cooperate or agree.
135 Spread of official information about taxes, elections, etc.
137 Radio communication.
140 Develop rural community spirit : destroy selfishness, iso
lationism.
The major fact about the communal social life of the village
is the presence of factions, political parties, etc.; these can be a
167
great hindrance to development (20, 129) . If the various communal
facilities of the village (5, 6, 24, 84, 89, 111, 115, 120, 137) are pro
vided in a central place, this place will very likely get associated
with one party, or certain families, and may actually not con
tribute to social life at all . On the other hand, it is important from
the point of view of social integration (21, 140) to provide a single
structure rather than isolated buildings. What is more, isolated
buildings also have the possible connection with the single family
nearest them, which can again discourage other families from going
there. What is required is a community center which somehow
manages to pull all the communal functions together so that none
are left isolated, but at the same time does not have a location more
in favor of some families than others. To achieve this, a linear
center, containing some buildings facing in, some out, zigzagging
between the different compounds, is necessary. This also meets
( 102) the need for processions with important stopping places ; and
adult literacy calls for a series of walls along the major pedes
trian paths, with the alphabet and messages written in such a
way that their continuing presence forces people to absorb them
( 1 16, 1 17, 135).
D1
>
..,
-
a:
,_
z
..,
APPENDIX 2 I MATHEMATICAL TREATMENT
OF DECOMPOSITION
which variables are referred to, we shall label the 1 ' s and O ' s with
subscripts. Thus p(Oi) is, specifically, the probability of xi taking
the value 0.
Consider M, or any of its subsystems S. Since each separate
variable takes the values 0 and 1 with equal probability, then if
the variables were all independent of one another, the 2m states
of M would be equiprobable, and for any S its 2• states would be
equiprobable. We should have :
q6
1 1
p (cr) = 2m for all cr, and p(A) = z. for all A .
condition :
p (OOOk)p ( l l Ok) - p(OlO k)p(lOO k )
v ,,.o .
=
[p( O ;O k) p ( 1 ;0 k) p( O ;O k) p ( 1 ;0k) ]�
I77
Among m variables, there are !m(m - 1 ) am-z such conditions
·
Condition 3
In any state of M, each of the m variables takes a fixed value.
Take any subsystem S. Suppose, without loss of generality, we re
number the variables so that x1 x. are in S, and x.+l
• • • Xm are · · ·
take the fixed pattern of values A, one for each possible pattern of
values taken by the set of m - s free variables x,+1 Xm . We
• • •
Condition 5
And we must have L P (<r) = I .7
(T
p(u) =
1 �!"o .
Take condition -1 first :
Now, if i and j are different, then in 2m-t cases e"; and e"; will take
the same sign so that their product is + 1 , and in �-l cases they
will take different signs so that their product is - 1 . Thus, for i
and j different, the sum over all 2m possible u, vanishes. For i and
j the same, V;; vanishes. Hence the last right hand side term is
identical to 0.
L P(u) 1 . =
"
variabJes
not in S
Since we get any subset S of M, by removing m - s variables
from M, one at a time, it is sufficient to prove the result for a
single step of removing one variable, and the general result follows
by induction. Consider, therefore, any variable Xk of M, and define
S as the subsystem obtained from M by removing xk . Pick an arbi
trary A. of this subsystem S. Suppose Ut and u 2 are the two states
of M in which the variables of S are in the same condition as in A.,
and for which xk takes the value 0 in u1 and the value 1 in U2.
1 80
We wish to prove that p (qi) + p (q2) =
p('A) .
To see that this is so we note that
=
;m (�IJ
�
V;;e",;e"d - 4
+
IJ
) v,;e"''e",; 2
I}
l'iie>.,exi ·
For i, j ;t. k , e11,;1 e11,; and e>.; are identical. For i or j k , the terms
=
from q1 cancel with those from q 2 , which makes the right-hand side
equal to 0 , and proves the point.
We now return to the correlation coefficients. Let us first take
the total correlation for a pair of variables, i and j. The above
result allows us to write the state probabilities of the two variable
subsystem, (x;,x i) , as :
p(OO) =
1 + V;jO p(10) =
1 - V;;O
4 4
p(01) =
1 - V;jO p( l l) =
1 + V;;O
4 4
where V; i is the number of links between x; and xi in G. This gives
a product moment correlation coefficient
p(OO) p(l l) - p(01)p( 10) = 4 V;;O i
16 4
j = V · ·O
[p(O)p(1)p(O)p(1) ]t '1 '
18 1
Suppo se the variables in S are held consta nt in some
fixed state A ,
we may then write
P(0 1 A) = 1 + ( k m + k x + k, - k;)o
'
2 -+2
P( 1 ·A)
1
= 1 + ( k x - k;) o .
2 -+1
The part ial correlati on is given by
4v;;O = V;;O
4(1 + 2k" o )
to the first order in o , which is very small. Hence the partial cor
relation is V;;O for all X, and satisfies condition 2.
Thus the measure
1 + k�o
p ( u) = --- 2"'
Xa
1 + 3o 1 - 30 1 - 0
p ( 0 1 1 1 ) = ----ul
--
p( OOOO ) =
--
16 p( OOl l ) =
16
1-- + 50 --0
1 - 1 - 3o
p ( 0001 ) = 1 6 p (0 1 0 1 ) = 16 p (1o l l ) = 1 6
p (OOlO) =
1
-
_.
.-
0
p ( 1001 ) =
1+ 0 1 - 0
--
p ( l 10 1 ) = 1 6
16 16
1 - 30 1+ 0 1 + 50
p( 0 1 00 ) =
--
p( 0 1 1 0) = 16 p ( 1 1 10) = 1 6
16
1--- 0 1 - 0 1 + 30
p( 1 000 ) = 1 6 p( 1 0 1 0 ) =
--
16 p(1 1 1 1) = 16
1 - 3o
p (l lOO ) = 16
Since we now have a workable probability distribution defined
over the states of M, we can write down an expression for the
average information carried by the system M. We use the Shannon
Wiener measure, and define H (M) , the average information carried
by M, as
- L P(u) log p(u).lo
"
We may rewrite this now, as
H(M) = - L mk"5 log
e � ) e �: ) "5
"
= - m L ( (1 + k" 5) [log ( 1 + k"5) - m log 2] }
; "
1 � {
= - m ( 1 + k"5) ( + k" 5 T k 252 + - m log 2) \f
2 �" - · · ·
{
= - 1m � - mlog 2+ ( 1 - m log 2)k"5 + T k 252
' terms in 53 .
+ and }
2 � .
above
In the sum, the constant term is counted 2m times . The term
in 5 vanishes, since we already know that L k" = 0. We there-
"
fore retain the term in 52, but drop the higher order terms, leaving
52
H (M) = m log 2 - m+l L k "2 •
2 "
Similarly we obtain, for any S,
52
H (S) = s log 2 - .+1 L k>-2 •
2 >-
We have defined ku =
L V;;eu;eui·
ii•L
Since we specified earlier that where there are several links be
tween a pair of vertices these links are individually identifiable,
we may now rewrite this expression as
ku = (2: eu;eu; - 2: eu;eu;) '
L+ L-
where each sum is taken over all the links belonging to L+ and L
respectively, so that the total contains l terms. It must be under
stood, of course, that this expression could be reduced, since each
of its l terms is either 1 or - 1. But, for the sake of clarity in the
following proof, we shall leave it in its expanded form. We may
write, then,
L (k u)2 =L (2:eu;eu; - L eu;eu;t
u L+ L-
=2: { (2: eu,eu;Y + (2: eu,eu;Y - z(2: eu,eu;2: eukeut) } .
u L+ L- L+ L-
the other half of the and is evenly distributed over the values
u,
r8s
taken by the e"h e"k ' and e" z ' we see that either of the above forms,
since they both contain an e"; raised to an odd power, will vanish
when summed over (f . Let us now look at the first and second
brackets in their expanded form. Again, all terms of the form
e" ; e" / e"k or e";e",.e"ke"1 will vanish when summed over a. There are
therefore only two kinds of term left, both of the form e" l e" ,-2 :
those which represent the same link taken_ twice, and those which
represent different links between the same vertex pair. We are
therefore left with
and similarly
Jlo
r 86
• s2
s1 • •
•
• •
• • • •
7r
s4
• • •
•
Sa
•
s• . s,, . . . M
where the sum r is taken only over pairs i,j, which are wholly
s.. s.. . . .
" partition-type." That is, the subsets it defines have s1,s2, Sp. · · ·
depend on the partition 7r. The first, l 0", is the number of the l0
potential spaces which are cut by the partition i . e . , the number
1r ,
"
that lo" � lo. The second of these parameters, l", is the number of
actual links cut by the partition 7r. This is given by l" L I V;; 1 . =
so this reduces to
-l-.. '
llo
'\' v ,; )
E(k =
.. 0
.. ..
We already know the value of the second term. As for the first:
E [ ( l: v ,; y] = E [ 2: v,;4 + 22: v1;2 v" 12 J .
.. .. ..
Since we have arranged to take V;; as positive, 0 or 1, we have =
� � � r
= p(v,;vkl 1) =
= .l l 1 - l(l - 1) .
lo lo - 1 lo(lo - 1)
=
= !lo.. (lo.. - 1 )
l(l - 1)
. lo(lo - 1 )
This gives us
Var ('\' v ;( ) = l-
·
- + 1
..
lo.. lo(l0 - l(l - 1 )
)
l l0.. 2 ( ) ·
';: l0 l o(lo 1 ) - T _
I89
llo"
102 . (l0 - 1 ) [lo2 - lo + lo(lo" - 1) (l - 1) - llo"(lo - 1)]
llo" "] - llo" ")
lo2(lo - 1) [l02 - l0l0 lo(lo - 1) (l0 - l 0 •
Again the variance depends on the value of lo" and hence on the
partition-type of 7r.
[Var (l")]!
and choose the constant to make this
lol" - llo"
[lo"(lo - lo")]!
This function has the same expected value and variance for all
partition-types, and may therefore be used to compare partitions
of all types with one another.
Expressed in terms of the earlier notation, this function is 16
pages 34-38 / 2 o o
tailed descriptive discussion see also H. von Foerster, " Basic Concepts
of Homeostasis," Homeostatic Mechanisms, B rookhaven Symposia in
Biology, No. 10 (Upton, N.Y., 1 957) , pp. 216-42.
28. This example is based on one given in Ashby, Design for a Brain,
p. 151.
29. Ibid.
30. See Chapter 9, note 4.
31. Ashby, pp. 1 92-204.
32. As Ashby puts it, "For the accumulation of adaptations to be
possible, the system must not be fully joined " (p. 155) .
33. This behavior of the misfits may be represented in step-function
form. See Ashby, pp. 87-90.
34. This would correspond to what Ashby calls ultrastability, ibid.,
pp. 122-37.
pages 38-47 / 2 o I
And the same is true of many other peoples. For instance: Gunnar
Landtman , " The Folk Tales of the Kiwai Papuans," A cta Societatis
Scientiarum Fennicae (Helsinki) , 47 (1917) : 1 1 6, and "Papuan Magic in
the Building of Houses," A cta A cademiae A boensis, Humaniora, 1
(1920) : 5.
8. Margaret Mead, A n Inquiry into the Question of Cultural Stability
in Polynesia, Columbia University Contributions to Anthropology, Vol.
9 (New York, 1928) , pp. 45, 50, 57, 68-69.
9. The blessing way rite, a collection of legends and prayers, makes
a positive link between their world view and the shape of the dwelling
by relating the parts of the hogan, fourfold , to the four points of the
compass, and by referring to them, always, in the order of the sun's
path-east, south, west, north. Thus one song describes the hogan's
structure : " A white bead pole in the east, a turquoise pole in the south,
an abalone pole in the west, a jet pole in the north ." The ritual involved
in the hogan's use goes further still, so far that it even gives details of
how ashes should be taken from the hogan fire. Berard Haile, " Some
Cultural Aspects of the Navaho Hogan," mimeographed, Dept. of An
thropology, University of Chicago, 1937, pp. 5-6, and " Why the Navaho
Hogan," Primitive Man, Vol. 15, Nos. 3-4 (1942) , pp. 41-42.
10. Hiroa Te Rangi (P. H . Buck) , Samoan Material Culture, Bernice
P. Bishop Museum Bulletin No. 75 (Honolulu, 1930) , p. 1 9 .
1 1 . L . G. Bark, " Beehive Dwellings of Apulia, " p. 409.
12. William Edwards, " To Build a Hut," The South Rhodesia Native
A ffairs Department A nnual (Salisbury, Rhodesia) , No. 6 ( 1 928) : 73-74.
1 3 . Iowerth C. Peate, The Welsh House, Honorary Society of Cymm-
rodorion (London, 1940) , pp. 183-90.
1 4. H . Frobenius, Oceanische Bautypen (Berlin, 1899) , p . 1 2 .
1 5 . Campbell, " Notes o n the Irish House," p. 223.
1 6. Clark Wissler, " Material Culture of the Blackfoot Indians,"
Anthropological Papers of the A merican Museum of History, Vol. 5,
part 1 (New York, 1 91 0) , p.·99.
17. L. G. Bark, " Beehive Dwellings of Apulia," p. 408 .
1 8 . A . I . Richards, " Huts and Hut-Building among the Bemba,"
Man, 50 (1950) : 89.
19. It is true that craftsmen do appear in certain cultures which we
should want to call unselfconscious (e.g., carpenters in the Marquesas ,
thatchers in South Wales) , but their effect is never more than partial.
They have no monopoly on skill, but simply do what they do rather
better than most other men . And while thatchers or carpenters may be
employed during the construction of the house, repairs are still undertaken
by the owner. The skills needed are universal, and at some level or other
practiced by everyone. Ralph Linton, Material Culture of the Marquesas,
pages 47-49 / 2 o 2
Bernice P. Bishop Museum Memoirs, Vol. 8., No. 5 (Honolulu, 1923) ,
p. 268. Peate, The Welsh House, pp. 201-5.
20. Barr Ferree, " Climatic Influence in Primitive Architecture,"
The A merican A nthropologist, 3 (1890) : 149.
21. Richard King, " On the Industrial Arts of the Esquimaux,"
Journal of the Ethnological Society of London, 1 (1848):281-82. Diamond
Jenness, Report of the Canadian Arctic Expedition (1913-1918), vol. 12:
The Life of the Copper Eskimos (Ottawa, 1922), p. 63 ; J. Gabus, " La
Construction des iglous chez les Padleirmiut," Bulletin de la Societe
Neuchateloise de Geographie, 47 (1939-40):43-51. D. B. Marsh, " Life in
a Snowhouse," Natural History, 60.2:66 (February 1951) .
22. W. G. Sumner, Folkways, p. 2.
23. Jenness, Copper Eskimos, p. 60.
24. W. McClintock, " The Blackfoot Tipi," Southwestern Museum
Leaflets, No. 5 (Los Angeles, 1936) , pp. 6-7.
25. Not only are the walls themselves daubed whenever they need to
be, but whole rooms are added and subtracted whenever the accommoda
tion is felt to be inadequate or superfluous. Meyer Fortes, The Web of
Kinship among the Tallensi (London, 1949), pp. 47-50. Jadk Goody,
" The Fission of Domestic Groups among the LoDagoba," ih The De
velopment Cycle in Domestic Groups, ed. J. Goody (Cambridge, 1958) ,
p. 80.
26. Whiffen, The North-West A mazons, p. 41.
27. Norbert Wiener, Cybernetics (New York, 1948), pp. 113-36.
28. Ibid., pp. 121-22 ; Ross Ashby, Design for a Brain (New York,
1960), pp. 100-4.
29. Strictly speaking, what we have shown concerns only the reaction
of the unselfconscious culture to misfit. We have not yet explained the
occurrence of good fit in the first place. But all we need to explain it,
now, is the inductive argument. We must assume that there was once a
very simple situation in which forms fitted well. Once this had occurred,
the tradition and directness of the unselfconscious system would have
maintained the fit over all later changes in culture.
Since the moment of accidental fit may have been in the remotest
prehistoric past, when the culture was in its infancy (and good fit an
easy matter on account of the culture's simplicity) , the assumption is
not a taxing one.
30. This is an obvious point. In another context Pericles put it
nicely: " Although only a few may originate a policy, we are all able to
judge it." Thucydides ii.41.
31. I am indebted to E. H. Gombrich for drawing my attention to
this phenomenon. The interpretation is mine.
pages 49-53 / 2 o 3
Chapter Five: The Selfconscious Process
1 . Thus selfconsciousness can arise as a natural outcome of scientific
and technological development, by imposition from a conquering culture,
by infiltration as in the underdeveloped countries today. See Bruno Snell,
The Discovery of the Mind, trans. T. G. Rosenmeyer (Cambridge, Mass.,
1 953) , chapter 10, " The Origin of Scientific Thought."
2. Hiroa Te Rangi (P. H. Buck), Samoan Material Culture, Bernice
P. Bishop Museum Bulletin No. 75 (Honolulu , 1930), pp. 85-86.
3. Ibid., p. 86.
4. For discussion of this development in present-day architecture see
Serge Chermayeff, " The Shape of Quality," Architecture Plus (Division
of Architecture, A. & M. College of Texas) , 2 (1959-60) : 16-23. For an
astute and comparatively early comment of this kind, see J. M. Richards,
" The Condition of Architecture, and the Principle of Anonymity, " in
Circle, ed. J. L. Martin, Ben Nicholson, and Naum Gabo (London, 1 937) ,
pp. 184-89.
5. In Chapter 3, an architecturally selfconscious culture was defined
as one in which the rules and precepts of design have been made explicit.
In Western Europe technical training of a formal kind began roundabout
the mid-fifth century B.C. And the architectural academies themselves
were introduced in the late Renaissance. Werner Jaeger, Paideia, Vol. I
(New York, 1 945), pp. 314-16; H. M. Colvin, A Biographical Dictionary
of English Architects, 1660-1840 (Cambridge, Mass., 1 954) , p. 1 6. It is
of course no accident that the first of these two periods coincided with
the prime of Plato's academy (the first establishment where intellectual
self-criticism was welcomed and invited) , and also with the first extensive
recognition of the architect as an individual with a name, and the second
with the first widespread crop of architectural treatises. F. M. Cornford,
Before and After Socrates (Cambridge, 1 932) ; Eduard Sekler, " Der
Architekt im Wandel der Zeiten," Der Aufbau, 1 4:486, 489 (December
1 959) .
6. For a detailed account of the origin and growth of the academies,
see the monograph by Nicolaus Pevsner, Academies of Art (Cambridge,
1940) , esp. pp. 1 -24, 243-95.
7. Margaret Mead, " Art and Reality," College Art Journal, 2:1 19
(May 1943) ; Ralph Linton, " Primitive Art," Kenyon Review, 3:42
(Winter 1941).
8. Ralph Linton, The Study of Man (New York, 1936), p. 3 1 1 .
\J. See Chapter 3 , pp. 41-42.
10. The invention and use of concepts seems to be common to most
human problem-solving behavior. Jerome Bruner et al., A Study of
Thinking (New York, 1956) , pp. 1 Q-17. For a description of this process
pages 57-61 / 2 o 4
as re-encoding, see George A. Miller, "The Magical Number Seven,
Plus or Minus Two: Some Limits on our Capacity for Processing Infor
mation," Psychological Review, 63 (1956):108.
11. See, for instance, American Association of State Highway Offi
cials, A Policy on Geometric Design of Rural Highways (Washington,
D.C., 1954), Contents; or F. R. S. Yorke, Specification (London, 1959),
p. 3; or E. E. Seelye, Specification and Costs, vol. II (New York, 1957),
pp. xv-xviii.
12. John Summerson, "The Case for a Theory of Modern Architec
ture," Royal Institute of British Architects Journal 64:307-11 (June 1957).
13. Serge Chermayeff and Christopher Alexander, Community and
Privacy (New York, 1963), pp. 159-175.
14. Reginald R. Isaacs, "The Neighborhood Theory: An Analysis of
Its Adequacy," Journal of the American Institute of Planners, 14.2:15-23
(Spring 1948).
15. For a complete treatment of this subject, see Rudolph Carnap,
Meaning and Necessity (Chicago, 1956). See esp. pp. 23-42, and for a
summary see pp. 202-4.
16. Ibid., p. 45.
17. It could be argued possibly that the word "acoustics" is not
arbitrary but corresponds to a clearly objective collection of require
ments- namely those which deal with auditory phenomena. But this
only serves to emphasize its arbitrariness. After all, what has the fact
that we happen to have ears got to do with the problem's causal struc
ture?
18. For the fullest treatment of the arbitrariness of language, as far
as its descriptions of the world are concerned, and the dependence of
such descriptions on the internal structure of the language, see B. L.
Whorf, "The Relation of Habitual Thought and Behavior to Language,"
in Language, Culture and Personality: Essays in Memory of Edward
Sapir, ed. Leslie Spier (Menasha, Wis., 1941), pp. 75-93.
19. L. Carmichael, H. P. Hogan, and A. A. Walter, "An Experimen
tal Study of the Effect of Language on the Reproduction of Visually
Perceived Form," Journal of Experimental Psychology, 15 (1932): 73-86.
20. Whorf, "Relation of Habitual Thought and Behavior to Lan
guage," p. 76. Whorf, who worked for a time as a fire insurance agent,
found that certain fires were started because workmen, though careful
with matches and cigarettes when they were near full gasoline drums,
became careless near empty ones. Actually the empty drums, containing
vapor, are more dangerous then the relatively inert full drums. But the
word "empty" carries with it the idea of safety, while the word "full"
seems to suggest pregnant danger. Thus the concepts "full" and "empty"
actually reverse the real structure of the situation, and hence lead to fire.
pages 6r-69 / 2 o 5
The effect of concepts on the structure of architectural problems is
much the same. Ibid., pp. 75-76. See also Ludwig Wittgenstein, The
Blue and Brown Books (Oxford, 1958) , pp. 1 7-20.
2 1 . Vitruvius, De architectura 3 . 1 , 3, 4. E. R. De Zurko, Origins of
Functionalist Theory (New York, 1 957) , pp. 26-28.
22. Werner Sombart, quoted in Intellectual and Cultural History of the
Western World, by Harry Elmer Barnes (New York, 1937), p. 509:
"Ideas of profit seeking and economic rationalism first became possible
with the invention of double entry book-keeping. Through this system
can be grasped but one thing-the increase in the amount of values
considered purely quantitatively. Whoever becomes immersed in double
entry book-keeping must forget all qualities of goods and services, aban
don the limitations of the need-covering principle, and be filled with the
single idea of profit; he may not think of boots and cargoes, of meal and
cotton, but only of amounts of values, increasing or diminishing. '� What
is more, these concepts even shut out requirements very close to the
center of the intended meaning! Thus in the case of"economics" even
such obvious misfit variables as the cost of maintenance and deprecia
tion have only recently been made the subject of architectural considera
tion. See J. C. Weston, "Economics of Building," Royal Institute of
British Architects Journal, 62: 256-57 (April 1955), 63: 268-78 (May 1 956),
63: 3 1 6-29 (June 1 956) . As for the cost of social overheads-the milk
man's rounds; the laundries and TB sanatoria which have to cope with the
effects of smoke from open fireplaces - even the economists are only just
beginning to consider these. See B enjamin Higgins, Economic Develop
ment (New York, 1 959), pp. 254-56, 66Q-61 . Yet the cost of the form is
found in all these things. The true cost of a form is much more compli
cated than the concept "economics" at first suggests.
pages 104-r09 / 2 1 2
16. For the isomorphism between dyadic relations and graphs see
Denes Konig, TheMie der endlichen und unendlichen Graphen (New York,
1950), pp. 107-9, and Claude Berge, Theorie des grapltes et ses applica
tions (Paris, 1958) , p. 6. Also for the isomorphism of dyadic relations and
square matrices see Irving M. Copilowish, "Matrix Developments of the
Calculus of Relations," Journal of Symbolic Logic, 1 3 : 193-203 (December
1948). For the extensional definition of a relation as the set of pairs related
under it, see Alfred Tarski, " On the Calculus of Relations," Journal of
Symbolic Logic, 6: 73-89 (March 1941).
17. In fact, as we shall see in Appendix 2, p. 187, the distinction
between positive and negative links is irrelevant, and we only need to
establish L, not L+ and L- separately. We shall also find it convenient
in practice to put JJ = 1, so that JJ;; can only be 0 or 1 .
18. I t i s sometimes quite hard t o draw the graph in a simple way,
so that the links are not all tangled. For a way to draw graphs, given
the matrix of links, see a recent paper published in the Journal of the
Acoustical Society of America, 33 (1961): 1183, on " Realization of a Linear
Graph Given Its Algebraic Specification."
19. See Appendix 2, p. 177.
20. See Appendix 2, p. 177.
21. See Appendix 2, p. 175.
'22. Let us note that this condition of e_qual " size" only refers to the
purely formal character of the system of variables. It does not imply that
the different variables have equal importance in the solution of the
problem. If it is more important to meet one requirement than another,
this still has no place in an analysis of the problem's causal structure,
but must be handled as it arises during the realization of the program.
23. We know that we shall never find requirements which are totally
independent. If we could, we could satisfy them one after the other,
without ever running into conflicts. The very problem of design springs
from the fact that this is not possible because of the field character of
the form-context interaction.
24. See the list of variables given in the worked example, Appendix
1, pp. 137-142.
pages r2s-r87 / 2 I 5
variate Correlation, " IBM Journal of Research and Development, 4 : 69
(January 1960) .
1 3 . See note 17 to Chapter 8.
14. Feller, Probability Theory, p . 2 1 3 .
15. T o normalize a random variable X, w e replace it b y (X - JJ.)/0',
where J1. is the mean and 0'2 the variance. See Feller, p. 2 1 5 .
1 6 . W e remember that lo = tm(m - 1) , l "
= L v;;, lo" L SaSp.
=
pages r88-r92 / 2 r 6
'